Abstract:During the process of drilling, the strong vibration and rapid rotation of bottom drilling tools make the attitude measurement signal contain multifrequency and highamplitude interference. The weak original signal amplitude and low signaltonoise ratio are difficult to be extracted in the field of measurement while drilling. To solve this problem, a Duffing chaotic oscillation detection method for weak signal recognition is proposed in this paper. Firstly, the frequency reconstruction of the measurement signal is realized by scale transformation. In this way, the measured signal can satisfy the restriction of frequency parameters. Then, to solve the influence of the initial phase of the measured signal on the accuracy of the detection model, Duffing oscillator detection model with full phase coverage is achieved by changing the initial phase of the driving signal. Finally, the threshold of Duffing oscillator entering the chaotic state is determined by adjusting the amplitude of the driving signal. The amplitude and phase parameters of the signal are estimated. The test results show that the root mean square error of inclination detected by chaos is 069%, and the relative error of the field-drilling is within ［107%, 208%］, which are higher than the results of original measurement data and standard Kalman filter. The feasibility and effectiveness of the proposed method has been proved.